This proposed pilot study will use existing data and key informant interviews across VA medical facilities to assess patient and facility factors associated with mental health care paid for by th VA, but provided outside of VA facilities (Fee Basis; also called non-VA care or contract care), and trends in growth of Fee Basis Care for mental health. Specifically, we will examine (a) variability among VA facilities in the proportion of Veterans receiving Fee Basis mental health care among those Veterans receiving mental health services within the VA, (b) whether the presence or absence of certain predisposing demographic or clinical characteristics explain within-facility variation in the relative mix or absolute use of Fee Basis vs. in-VA mental health care among those Veterans receiving mental health services within the VA, and (c) facility-level characteristics associated with growth trends and deviations from expected use for Fee Basis mental health care from fiscal years 2002 to 2013 as a proportion of all mental health care delivered. We will use existing administrative data comprising electronic medical records for Veterans (outpatient, inpatient, and residential), sociodemographic information, and facility-level characteristics to assess variation in Fee Basis mental health care among a sample of roughly 2.5-3.0 million Veterans receiving their treatment in the VA medical system between fiscal years 2002 and 2013. We will extend a model of health service utilization to characterize and quantify individual- and facility-level factors that explain the extent to which Fee Basis-only, in-VA only, or mixed in-VA/Fee Basis mental health services are used, as well as trends in growth of purchased care for mental health. We will limit analyses to posttraumatic stress disorder, depression, and substance use disorders as these represent 1) the three most prevalent psychiatric conditions for which Veterans receive any VA mental health services in the VA; 2) the most prevalent diagnoses for Fee Basis mental health care; and 3) conditions that co-occur in many VA patients. To identify variables not accounted for in administrative data that may affect Fee Basis Care utilization, we will conduct semi- structured interviews with 2-3 VA mental health administrators from each of the six facilities that represent the three highest and three lowest users of Fee Basis Care. Semi-structured interview guidelines will be developed based on key variables and patterns from quantitative analyses, and will include open-ended questions to capture unique facility characteristics. These results will provide the VA with immediate information about the Fee Program with respect to mental health services and can inform resource allocation planning efforts for current and future VA mental health funding. Furthermore, this pilot study will lay the groundwork to identify circumstances and locations in which provision rather than procurement of mental health services yields better outcomes for both the Veteran and the VA and improves health care value. Decisions surrounding the best value for mental health services require an understanding of complex factors. By examining variability in mental health care among Veterans based on care setting, the VA may better identify potential health care system gaps and opportunities to meet the treatment needs for Veterans diagnosed with mental health conditions.